I had a post at Wonkblog over the weekend, “Is a democratic surveillance state possible?”
In some sense, the issue of the government spying and collecting data on its citizens isn’t a new problem. One of my favorite tweets of the past week was Brooke Jarvis noting "Collapsing bridges alongside massive spy networks... Ah, the Jeffersonian ideal of government."
The United States has been tracking, observing, and surveilling its citizens for centuries. That includes that long-standing form of communication, the mail. As Senator Lindsey Graham just said, “In World War II... you wrote a letter overseas, it got censored...If I thought censoring the mail was necessary, I would suggest it.”
From the Census in the Constitution to the Cold War spy network (including the NSA, founded in 1952 through the Executive Branch), maybe this should be seen as a continuation of an old issue rather than a brand new one. But I think there are genuinely new and interesting problems with the 21st century Surveillance State and the brand new digital technologies that create the foundations for it. What’s new about the new surveillance state?
1. It’s always on and always has been. Old acts of surveillance had to be triggered and were forward-looking. However, we now spend so much of our lives online, and that is always being recorded. As the leaker Edward Snowden said in his interview, “they can use this system to go back in time and scrutinize every decision you've ever made, every friend you've ever discussed something with. And attack you on that basis to sort to derive suspicion from an innocent life and paint anyone in the context of a wrongdoer."
To the extent that old surveillance was capable of going back, it was by checking old records or interrogating old sources. And there the concept of amnesia comes into play.
2. It will never forget. “Amnesia” is a normal front line of defense. People forget things. Clear memories, stories, and ideas become grey. Photos and documents get lost with time. Trying to piece together history will necessarily involve a lot of missing gaps and poor recollection.
Not with the surveillance state. Cheap digital storage means that clear, easily replicable data will exist for the foreseeable future.
3. It scales easily. If the FBI was keeping records on 100 people in the 1950s, and it then wanted to monitor 1,000 people, it would probably need 10 times as many resources. Certainly it wouldn’t be effortless to scale up that level of surveillance.
As we can see in the age of Big Data and fast computing, this is no longer the case. The resource costs of accessing your phone’s metadata history versus all phones’ metadata history is going to be (somewhat) trivial. And the fact that there’s no amnesia means that you’ll always have access to that extra data.
4. It’s designed to be accessible. As Orin Kerr emphasizes, digital data here isn’t collected or surveilled via the human senses. A person can’t simply “peek” into your email the way they could peek at your physical mail. Instead devices need to be installed to access and make sense of this data. Private sector agents will do this, because it is part of their business model to make this information accessible. These access points will also be accessible to government agents under certain conditions - part of the major debate over the PRISM program is under what conditions the government can actually access these devices.
5. It’s primarily driven by the private sector. Broadly speaking, measures of democratic accountability and constitutional protections do not extend to the private sector. More on this soon, but things like the Freedom of Information Act to the Administrative Procedure Act to our whole regime of transparency laws do not apply to outside businesses. The government has worked with private groups before on surveillance, but here it is in large part driven by private agents, both for contractors and information gathering.
6. It predicts the future for individuals using mass data. Surveillance has generally used mass data to either predict or determine future courses of action on a mass scale. For instance, Census data is used to allocate federal money, or predict population growth. Alternatively, it uses individual data to analyze individual behavior - asking around and snooping to dig up dirt on someone, for instance.
The surveillance state, however, allows for using mass data to predict the actions of individuals and groups of individuals. This is what generates your Netflix and Amazon suggestions, but it is also now providing the basis for government actions. As Kieran Healy notes, this would have been interesting back in the American Revolution.
This is distinct from the normal Seeing Like a State (SLS) critique of how states see their citizens. Some think states produce “a logic of homogenization and the virtual elimination of local knowledge...an agency of homogenization, uniformity, grids and heroic simplification” (SLS 302, 8). But rather than flatten or homogenize its citizens when observed under bulk conditions, it actually creates a remarkably individualized image of what its citizens are up to.
What else is missing, or shouldn't have been listed? You could view these as a technological evolution of what was already in place, and in some ways that would make sense. But the technology has opened a brand new field. This existed before the War on Terror, and will likely exist afterwards; dealing with the laws and institutions behind this new state is crucial. As the technology has changed, so must our laws.
Follow or contact the Rortybomb blog:
I had a post at Wonkblog over the weekend, “Is a democratic surveillance state possible?”
In some sense, the issue of the government spying and collecting data on its citizens isn’t a new problem. One of my favorite tweets of the past week was Brooke Jarvis noting "Collapsing bridges alongside massive spy networks... Ah, the Jeffersonian ideal of government."
The United States has been tracking, observing, and surveilling its citizens for centuries. That includes that long-standing form of communication, the mail. As Senator Lindsey Graham just said, “In World War II... you wrote a letter overseas, it got censored...If I thought censoring the mail was necessary, I would suggest it.”
From the Census in the Constitution to the Cold War spy network (including the NSA, founded in 1952 through the Executive Branch), maybe this should be seen as a continuation of an old issue rather than a brand new one. But I think there are genuinely new and interesting problems with the 21st century Surveillance State and the brand new digital technologies that create the foundations for it. What’s new about the new surveillance state?
1. It’s always on and always has been. Old acts of surveillance had to be triggered and were forward-looking. However, we now spend so much of our lives online, and that is always being recorded. As the leaker Edward Snowden said in his interview, “they can use this system to go back in time and scrutinize every decision you've ever made, every friend you've ever discussed something with. And attack you on that basis to sort to derive suspicion from an innocent life and paint anyone in the context of a wrongdoer."
To the extent that old surveillance was capable of going back, it was by checking old records or interrogating old sources. And there the concept of amnesia comes into play.
2. It will never forget. “Amnesia” is a normal front line of defense. People forget things. Clear memories, stories, and ideas become grey. Photos and documents get lost with time. Trying to piece together history will necessarily involve a lot of missing gaps and poor recollection.
Not with the surveillance state. Cheap digital storage means that clear, easily replicable data will exist for the foreseeable future.
3. It scales easily. If the FBI was keeping records on 100 people in the 1950s, and it then wanted to monitor 1,000 people, it would probably need 10 times as many resources. Certainly it wouldn’t be effortless to scale up that level of surveillance.
As we can see in the age of Big Data and fast computing, this is no longer the case. The resource costs of accessing your phone’s metadata history versus all phones’ metadata history is going to be (somewhat) trivial. And the fact that there’s no amnesia means that you’ll always have access to that extra data.
4. It’s designed to be accessible. As Orin Kerr emphasizes, digital data here isn’t collected or surveilled via the human senses. A person can’t simply “peek” into your email the way they could peek at your physical mail. Instead devices need to be installed to access and make sense of this data. Private sector agents will do this, because it is part of their business model to make this information accessible. These access points will also be accessible to government agents under certain conditions - part of the major debate over the PRISM program is under what conditions the government can actually access these devices.
5. It’s primarily driven by the private sector. Broadly speaking, measures of democratic accountability and constitutional protections do not extend to the private sector. More on this soon, but things like the Freedom of Information Act to the Administrative Procedure Act to our whole regime of transparency laws do not apply to outside businesses. The government has worked with private groups before on surveillance, but here it is in large part driven by private agents, both for contractors and information gathering.
6. It predicts the future for individuals using mass data. Surveillance has generally used mass data to either predict or determine future courses of action on a mass scale. For instance, Census data is used to allocate federal money, or predict population growth. Alternatively, it uses individual data to analyze individual behavior - asking around and snooping to dig up dirt on someone, for instance.
The surveillance state, however, allows for using mass data to predict the actions of individuals and groups of individuals. This is what generates your Netflix and Amazon suggestions, but it is also now providing the basis for government actions. As Kieran Healy notes, this would have been interesting back in the American Revolution.
This is distinct from the normal Seeing Like a State (SLS) critique of how states see their citizens. Some think states produce “a logic of homogenization and the virtual elimination of local knowledge...an agency of homogenization, uniformity, grids and heroic simplification” (SLS 302, 8). But rather than flatten or homogenize its citizens when observed under bulk conditions, it actually creates a remarkably individualized image of what its citizens are up to.
What else is missing, or shouldn't have been listed? You could view these as a technological evolution of what was already in place, and in some ways that would make sense. But the technology has opened a brand new field. This existed before the War on Terror, and will likely exist afterwards; dealing with the laws and institutions behind this new state is crucial. As the technology has changed, so must our laws.
Follow or contact the Rortybomb blog:






